goliathDown

How are you handling shading for dense wooded areas? I'm looking around at some neighborhoods with dense redwood trees in the northern California region. I know in the middle of the day a decent amount of shadow will still be in the area, which comes from the density of the forest and the length of the trees. From what I'm seeing, this shading is not accurate to the wooded areas I frequent.

But, that's probably a really hard problem to tackle. If there's no data on tree height, it seems impossible to accurately portray shadow extrapolations for forests. Especially since the forests can have a high frequency of change.

Super cool project, I hope this continues to grow!

Really interesting article, excited to see the followup!

A few things that I think the author could take into consideration. What is a country's population actively discussing versus other countries? Is there a factor between direct and indirect exposure of problems that may influence a person's stance?

Even if there is zero evidence supporting an influence campaign on the platform, the ease of collecting user data or spying on users is something I would expect an active adversary to do. Like it or not, China and America are at odds with each other, and it's almost silly to assume that China would not be exploiting a successful tool for their own means.

What type of company does China keep? Last I checked, they were pretty buddy-buddy with Russia, and North korea on an estranged leash. If we know other countries fund terrorist organizations for intent of undermining western influence, is it that big of a stretch to say China is probably doing the same thing?

You can't use 'direct conflict' as a measure in this type of game. Russia has done damning harm to american politics, but its not clear how to measure that effect.

That's a bit of a ridiculous standard. Mostly because I don't think China is liable to hand out the records they've been gathering to cross check findings from other study.

Isn't China THE foreign adversary though? I think there's a difference between a media company housed in the UK or France versus China or Russia.

/dads/

Somethin about this just makes me smile. Nice site

I wish the editor (or team) did a better job in proofreading their article. The multiple spelling mistakes and grammar mistakes makes me question the credibility of their ability to report.

I feel like that's not the case

Just skimming to two parts in the second to last paragraph, these two sentences, "I am often desperately looking inward to find the strength to go on", and "Accepting a constant stream of low quality things for most of my life, including interactions with other people, only contributed to the way my life has gone and the way I feel now" were really striking to the author's day-to-day. It sounds like he' struggling to doing what he needs to do and they're content with the way this is. We should want more for ourselves and for our relationships to have meaning.

I'm curious as to what makes a good size of N. To me, N=10 would be do small, N=48 seems a little under what is should be, and N=100 seems sufficient, but I don't have a real basis for this (in fact, it's probably because of the original count that had me settle on a min and max of 10 and 100). Maybe a better question is: What are factors that studies consider, besides the logistics you rightly pointed out, in determining a sufficient N?

300 series are advanced graduate classes. As an undergrad the sight of a 300 is horrifying.

I'm relatively new to ReactJS and professional web development, and I've been using (and learning) react-redux. The layer schema makes sense to me; React handles user input, redux manages state, and react-redux ties them together.

If react does a better job at speedily handling user input to the point of seamless interaction, then I suppose that would be good. What are we losing from choosing to make a layer itself responsible for the speed of returning output (rendering) from input (user interactions)?

This was for Computer Vision and NLP conferences, but would the same be true if AutoML were thrown into the mix? I care mostly about efficiency and optimization, and the author wasn't able to distinguish that Pytorch is any better or worse save the two anecdotes.